Plant galls recorded from Guanacaste Conservation Area-Costa Rica as an integrated concept of a biological database

Galling insects are specialist herbivorous that have the ability of manipulating plant tissue to form complex biological structures called galls. Even though different organisms have the ability to induce galls in plants, insect galls have the highest degree of structural complexity. The main goal of this study was to obtain a preliminary systematic record of plant gall morphotypes from the Guanacaste Conservation Area in Costa Rica and integrate the information into a biological database. Plant gall morphotypes were recorded, characterized and deposited into a specialized herbarium established as a reference for the inventory. Moreover, organisms associated with gall morphotypes were included in the inventory when it was possible to obtain and identify them. Galls were collected in the rainy season over a period of three years. In total, we recorded forty-four families, seventy genera, and eighty-seven host plant species. One hundred thirty-one morphotypes of plant galls were identified in the Guanacaste Conservation Area. The family with the highest number of gall morphotypes was Fabaceae (8.4%). Leaves were the organ with the largest number of galls (71%), followed by stems (17.6%), and apical buds (6.9%). The predominant gall shape was globular (25.2%), followed by discoid (18.3%). Fifty-nine percent of the galls had a glabrous texture, which was most common on leaves, with 77%. One hundred twenty of our field records (91.6%) of plant galls were new morphotypes not only for Costa Rica but also the world. As a consequence of this research and considering the prospect of future increases in new gall records (and associated organisms), we proposed having the biological entities resulting from the inventory placed in a cecidiarium. This repository represents a standardized and comprehensive way to manage the data and biological materials associated with the plant galls. We also suggest a nomenclature for standardizing gall morphotype registries and identifications. This work is the first and most detailed inventory of plant galls carried out thus far in the Guanacaste Conservation Area.


Introduction
Plant galls are atypical plant tissue structures induced by the action and activity of a foreign organism. Although several organisms have the ability to induce galls in plants, the most diverse and complex galls are formed by insects (Shorthouse & Rohfritsch 1992). Gallinducing insects are highly specialized sedentary herbivores, which feed specifically on certain specialized cells that are found within the plant structure whose formation they have induced (Shorthouse & Rohfritsch 1992, Tooker et al. 2008, Raman 2011. Gall-inducing insects generally have specific host plants (Cuevas-Reyes et al. 2014). Gall biology is closely associated with the respective inducing insect, in such a way that galls induced by a particular insect species are basically always the same shape and can differ distinctly from others induced by related species. On the other hand, there are some gall morphotypes induced by different species of insects that present similar shapes (Raman 2011). Distinctive characteristics of each type of gall are probably due to slight variations in the way that each insect species stimulates the development of the gall in the corresponding plant tissue (Shorthouse & Rohfritsch 1992). Li et al. (2017) suggested that gall development is influenced by the gall-inducing insects as well as by the tissue developmental stage and plant genotypes.
The use of gall morphotypes is a commonly used and reliable parameter because evidence indicates that each gall is unique to a particular gall-inducing insect (Stone and Schönrogge 2003), and each galling species is specific to a particular host plant (Abrahamson et al. 1998). According to Isaias et al. (2013), a gall morphotype could be defined as a characteristic phenotypic variation in a neo-formed plant organ, which is produced by the species-specific interaction between the inducing organism and a specific host plant. Due to the fact that each gall inducer is able to alter the morphogenesis in a predetermined organ (Rohfritsch 1992), and because specificity of galler taxa is strongly linked to an appropriate oviposition site in the host plant (Eigenbrode & Jetter 2002), gall morphotype is widely used as a way to refer to types of plant galls. The vast majority of gall-inducing arthropods are restricted to a single host plant species, thus corroborating the idea that the gall morphotype can be used as reliable substitute of gall-inducing species. In addition, gall polymorphism, which could lead to failures in the identification of galls, appears to be a rather rare phenomenon ).
Thousands of gall-inducing insects have been identified around the world, for the most part belonging to the orders Thysanoptera, Hemiptera, Lepidoptera, Coleoptera, Diptera, and Hymenoptera (Rohfritsch & Shorthouse 1982, Leitch 1994, Williams 1994, Hanson & Gómez-Laurito 2005, Espírito-Santo & Fernandes 2007, Ansaloni et al. 2018. Gall-inducing insects induce the formation of galls in leaves, buds, flowers, stems, roots and other organs of the plant, depending on the gall-inducing species. A growing number of studies on the diversity and abundance of plant gall morphotypes, and their respective inducing insects, allowed us to build upon the existing information. With a certain regularity reports are made of records and descriptions of new species of gall-inducing insects (Ronquist & Liljeblad 2001, Dalbem & Mendonça 2006, Güçlü et al. 2008, Coelho et al. 2009, Maia & Oliveira 2010, Maia et al. 2010a, Maia et al. 2010b, Medianero et al. 2010, Sano et al. 2011, Santos et al. 2011a, Maia 2014, De Araújo 2017, Bergamini et al. 2017, Coelho et al., 2017, Martins et al. 2018, Ley-López et al. 2019. Inventories of biological specimens are an important tool to know and preserve biological diversity. To date, for tropical regions, detailed inventories on the occurrence and diversity of gall morphotypes are scarce and incomplete. In the Neotropical Region, the majority of plant gall morphotype records are from Brazil (Urso-Guimarães & Scarelli- Santos 2006, Saito & Urso-Guimarães 2012, Isaias et al. 2013, Carvalho-Fernandes et al. 2016, Urso-Guimarães et al. 2017, Araújo 2017, Liu et al. 2018, Flor et al. 2018, Ascendino & Maia 2018, Vieira et al. 2018, Ribeiro et al. 2019 and some of them were conducted in different types of tropical dry forests or semi-arid ecosystems (Coelho et al. 2009, Santos et al. 2011b, Maia & Souza 2013, Costa & Araújo 2019. Likewise, other reports have been done for the Mexican dry forest (Cuevas-Reyes et al. 2004, Cuevas-Reyes et al. 2014. For the Costa Rican flora there are no systematic and detailed records of plant galls for specific regions or ecosystems. Studies are restricted to sporadic or general reports (Gómez & Kisimova-Horovitz 1997, Hanson & Gómez-Laurito 2005, Retana-Salazar & Nishida 2007, Retana-Salazar & Sánchez-Chacón 2009. Hanson et al. (2014) reported 1,100 morphospecies of gall-inducers, distributed in six orders of insects, from Costa Rica. Nevertheless, this survey was restricted to the identification of the insect gall inducers and their associated parasitoids, using plant gall morphotypes as indirect indicators of insect morphospecies, and therefore lacking detailed descriptions of the gall morphotypes found, usually without associated images. The work carried out by Ley-López et al. (2019) in the area of Sarapiquí, in northern Costa Rica provides primarily a checklist of the native vascular plants found to harbor galls.
The Guanacaste Conservation Area (Área de Conservación Guanacaste, ACG for its acronym in Spanish) is located in the northwest region of Costa Rica, in the life zone categorized as tropical dry forest (Holdridge & Tosi 1967). ACG is a conglomeration of several national parks and other natural areas (sectors), among which are Santa Rosa National Park, Guanacaste National Park, Rincón de la Vieja National Park, Murciélago Sector, among others (SINAC 2014). ACG protects the most emblematic tropical dry forest in Central America (Quirós-Arias 2017). This Costa Rican national park was created to favor the restoration process of old pasture areas into the primary and secondary forest through natural succession. Since 1999 it it has been proclaimed a World Heritage Site by UNESCO (Quirós-Arias 2017).
The present work is a first attempt to establish a detailed record and inventory of plant gall morphotypes and their associated organisms in the Guanacaste Conservation Area utilizing an integrated approach consisting of a biological database with a specialized herbarium (cecidiarium).

Study area
Fieldwork and sample collection were carried out predominantly in the Pacific area of Guanacaste Conservation Area (Área de Conservación Guanacaste, ACG), Guanacaste province, Costa Rica, located in the northwest region of the country (10.839366, -85.618176, administrative area) ( Figure 1A). Plants and gall morphotypes were collected in different ecosystems in ACG. The Guanacaste Conservation Area is characterized by having a long dry season, generally from late November to mid-May, with a rainy season from mid-May to November (Borchert 1994, Lobo-Segura 2019. The sampled sites are located in the tropical dry forest ecosystem, as well as areas in the premontane wet transition, according to the Holdridge & Tosi (1967) classification. In the last decades, ACG has been affected by long periods of drought, being one of the areas most affected by the El Niño phenomenon and climate change projections (Quirós-Arias 2017).
The Guanacaste Conservation Area includes 296,526 acres on land and 106,255 acres of marine habitats. Approximately 65% of all Costa Rican species are present in this area, representing 2.6% of the world's biodiversity (SINAC 2014).

Field and laboratory work for gall morphotypes inventory and sampling associated insects
Galls were sampled by randomly walking the trails and examining the vegetation along the pathways by two people searching simultaneously on both sides of the road. All plants and their aerial organs were observed to a maximum height of approximately 3 meters. Each gall and host plants found were collected and then packed in plastic bags for further processing. The photographic record of each gall morphotype, when possible, was made directly in the field. Randomized trials that were perpendicular to existing trails were used occasionally for sampling galls, with distances varying from 50 to 400 meters, but without a pre-established experimental design.
Plant galls were collected and processed from May 2010 to December 2012 in different types of vegetation that comprised the ACG biome. Field sampling was carried out every 3 or 4 months per collection season, for a period of three days each, covering approximately 6-8 hours of field work per day. Sampling was conducted especially during the rainy season, when there is a greater probability of finding leaves on the plants, due to the marked dry season and a predominantly deciduous forest. Gall samples were collected mainly at lower-medium altitudes of 0 to 1100 m above sea level (m.a.s.l.): Santa Rosa Sector (La Casona area), Murciélago Sector, Santa Elena Sector, Pocosol Sector (Góngora Area, road to Cacao Volcano), and Rincón de la Vieja National Park. Moreover, gall samples and their corresponding host plants were collected mostly from deciduous plant species typical of non-riparian environments as well as near sites with original-natural forest with evergreen species. More typical riparian sites or similar environments followed comparable field zonification criteria of Frankie et al. (1974) and Borchert (1994). Two Bosque Viejo (Old Forest) patches were sampled: close to La Casona (near the administrative area of Santa Rosa National Park) and from the route to Playa Naranjo Sector. The Bosque Viejo patches are considered remnants of the original forest with little human disturbance and is characterized by an abundance of evergreen plant species.
Specimens of each plant harboring galls were prepared for reference as herbarium vouchers to be identified later. Plant gall specimens with floral or fruit parts for host identification were dried when possible, otherwise the insects obtained, and relevant gall samples were preserved in 70% ethanol. Since gall traits changed in dried samples, especially their morphology and color, a printed photographic registry of each specimen is included in the inventory. All samples were deposited at the cecidiarium (specialized herbarium) established at the facilities of Technological Institute of Costa Rica (Instituto Tecnológico de Costa Rica, Campus Tecnológico Local San Carlos, Escuela de Ciencias Naturales y Exactas). Also, some selected gall samples were preserved in FAA (formalin, acetic acid, alcohol) for later morphological studies. Collected galls were classified according to place of origin, family, genus, and species of the host plant, and according to the basic morphological type. Sampling site description was registered for all samples, but when there were more than two accessions of plant materials in the inventory, only one description of the sampling site for each morphotype was included here. Collected samples were georeferenced by Global Positioning System (GPS). Each gall morphotype location was geo-referenced using the Decimal Degrees (DD) format. A map to show the reference morphotypes collected per site at ACG was made with QGIS version 3.10.5 (QGIS.org 2020) Development Team and using the following layers from Atlas Digital CR 2008SINAC 1998) To facilitate visualization of geographical location, some neighboring galls collected from nearby areas (i.e., spaced between 5 m to 400 m linear meters) were considered as a cluster, depending on the geographic characteristics of the site; for these clusters the same geographic position was assigned. Park, and Rincón de la Vieja National Park sectors. Taken and modified from Área de Conservación Guanacaste-SINAC (https://www.acguanacaste.ac.cr/ index.php), and Google Earth®. 1B) Plant Gall Herbarium (cecidiarium) organization. 1) Dried plant gall specimens, 2) Plant gall samples preserved in 70% ethanol and FAA, 3) Insect collection of gall inducers, parasitoids, and inquilines preserved in 70% ethanol vials, 4) Digital database containing information about gall morphotypes, host plants, gall makers, associated organisms, sampling sites, etc.
In addition to the list of host plants harboring galls, the morphological characterization of all gall morphotypes found is provided. Gall morphotypes were used to register plant galls, as well as a reference for the galling insect collection, assuming that each morphotype is unique and each galling species is specific to its host plant (Espírito-Santo & Fernandez 2007, Abrahamson et al. 1998). The morphological characteristics used in gall differentiation were basic gall form, external color and epidermial structure. Morphotype shape was established according to the most usual morphological classifications found in the literature on this topic (Maia 2001, Urso-Guimarães et al. 2003, Santos et al. 2011a, Saito & Urso-Guimarães 2012, Isaias et al. 2013. Since these classification criteria could be ambiguous, we classified galls based on their basic general shapes considering that gall morphology could be a mixture of different shapes.
Gall morphotypes were named according to the two first letters from the binomial scientific name, followed by the numerical order of appearance in each plant species, and if necessary, the third letter of the specific epithet might also be used. We propose this nomenclature designation to avoid confusion in the registry of gall morphotypes. Moreover, only detailed literature descriptions with photographs were considered for previous records of plant gall morphotypes.
Adult stages of the gall-inducers and their parasitoids were obtained by rearing galls in plastic bags until the adult emerged. Gall inducers, parasitoids, and inquilines were preserved in plastic vials containing 70% ethanol and deposited in the cecidiarium. Roberto Espinoza carried out the taxonomic identification of the host plants, and the inductor insects were identified by Paul Hanson as much as possible.
Data from external gall epidermis lignification and trichomecovered galls from deciduous forest areas and evergreen tree areas such as Bosque Viejo were statistically analyzed by a Chi-square test.

Gall morphotype inventory and collection of associated organisms in a specialized herbarium as an integrated biological database
Inventories of biological specimens are a valuable tool to know and preserve biological diversity; for plant specimens, for instance, a traditional herbarium is an appropriate way to carry out this task. A specialized herbarium of plant galls was created to become a reference collection for Costa Rican plant galls. This herbarium began operating in 2012 and currently has around 400 sample accessions. The collection started with plant galls from the Guanacaste Conservation Area. To date, the herbarium is made up of four basic units: the plant gall collection of dried specimens, selected gall samples preserved in 70% ethanol (some of them stored at -70ºC), associated insects preserved in 70% ethanol, and a digital database with all the collected information ( Figure  1B). Due to the fact that gall traits change in dried samples, a printed photographic registry of each specimen is included in the collection, and for further morphological studies, a collection of selected galls in FAA (formalin-acetic acid-alcohol) is expected to be included in the future. A database with plant gall data and photographs, as well as information related to their associated organisms, is expected to be available using FileMaker-Pro software or another similar program. This specialized herbarium functions according to appropriated technical standard and collections are maintained in a controlled environment at 20 C° with relative humidity between 40-60%.

Results
A total of eighty-seven species, in seventy genera and forty-four families, of plants that host galls were recorded in the Guanacaste Conservation Area (Table 1). We found one hundred thirty-one morphologically distinct types of plant galls in ACG. The plant families with the highest number of gall morphotypes were Fabaceae (8.4%), Rubiaceae (7.6%), Malvaceae (6.1%), Sapindaceae (5.3%), Boraginaceae (4.6%), and Nyctaginaceae with 4.6% (Table 2). Sixty plant species harbored one gall morphotype, fifteen had two associated morphotypes, six plant species harbored three gall morphotypes, two species contained four gall morphotypes and three species harbored five morphotypes. The species with the greatest number of galls were Acalypha diversifolia (Euphorbiaceae) and Psychotria horizontalis (Rubiaceae) with four gall morphotypes, as well as Pisonia macranthocarpa (Nyctaginaceae), Sideroxylon obtusifolium (Sapotaceae), and Stegnosperma cubense (Stegnospermataceae) with five morphotypes each.
Some gall samples were so rare that there were not enough to obtain insects, but in many cases, although we had enough plant material, it was not possible to obtain adult insects for identification purposes. Gall-inducer identification to family level was possible in many cases based on the larval stages encountered during the dissections of some selected galls when enough material was available. Nevertheless, even when adult stages were obtained, identification of most insects beyond the family level was complicated by the lack of appropriate taxonomic references, a limitation described by Hanson et al. (2014). Therefore, most of the collected insects remain as unidentified species. The inducing insects that were identified belong to the family Cecidomyiidae (Diptera). Some parasitoids/inquilines (all belonging to the order Hymenoptera) were identified to the family, subfamily or genus level.
Plant gall morphotype description, name and characteristics are presented below under host plant families, genera and species in alphabetical order. They included gall morphology classification, color, epidermial structure, organs attacked, associated organisms as well as host plant description, location and geographical coordinates. Anacardiaceae Astronium graveolens Jacq. Morphotype As_gr_1 ( Figure 2C). Gall description: Globular shape, yellowish green to brown at maturity, induced on leaves, lignified epidermis. Glabrous, on the adaxial surface of the leaves. Gall inducer: unknown. Parasitoids/Inquilines: unknown. Plant description: Sapling, nearly 5 m tall, barren. Location: Guanacaste, Liberia, Nacascolo. Santa Rosa National Park, Bosque Viejo Area between the entrance of the park and La Casona. Coordinates/Altitude: 10,85072 N 85,60796 W, 321 m. Registry comments: Gall recorded by Hanson & Nishida (2014).
Parasitoids/Inquilines: unknown. Plant description: Vine, barren. Bifoliate leaves with cuspidate apex. Location: Guanacaste, Liberia, Nacascolo. Santa Rosa National Park, Bosque Viejo Area between the entrance of the park and La Casona. Coordinates/Altitude: 10,85072 N 85,60796 W, 321 m; 10,83581 N 85,62347 W, 306 m. Registry comments: First gall morphotype record for Guanacaste Conservation Area, Costa Rica, and the world. However, one gall induced by Cecidomyiidae on this plant species was reported, without a reference image, by Medianero et al. 2010.
Erythroxylum macrophyllum Cav. Morphotype Er_ma_2 ( Figure  3H). Gall description: Spherical shape, white-yellow, induced on the upper and lower surface of leaves, glabrous epidermis. Gall inducer: unknown. Parasitoids/Inquilines: unknown. Plant description: Shrub, nearly 2m tall, barren. Leaves with acute apex and entire margin. Location: Guanacaste, Liberia, Curubandé. Guanacaste National Park, Las Pailas Area, in the old secondary forest after the pasture, on the way to the crater. Coordinates/Altitude: 10,78427778 N 85,3484167 W, 955 m. Registry comments: First gall morphotype record for Guanacaste Conservation Area, Costa Rica, and the world.
Guazuma ulmifolia Lam. Morphotype Gu_ul_2 ( Figure 5E). Gall description: Irregular shape, green and yellow, distributed on the glabrous adaxial surface of the leaf, glabrous epidermis. Gall inducer: unknown. Parasitoids/Inquilines: unknown. Plant description: Sapling, nearly 3 m tall, barren. Ovate leaves with dentate margin. Location: Guanacaste, Liberia, Mayorga. Guanacaste Conservation Area, Góngora, roadside to the Cacao Volcano. Coordinates/Altitude: 10,88683333 N 85,47311111 W, 597 m. Registry comments: First gall morphotype record for Guanacaste Conservation Area, Costa Rica, and the world, although a gall on the same plant organ for this plant species was reported, without a reference image, by Coelho et al. (2014) but gall description doesn't match.
Except for 11 records, 120 of all recorded galls are new records for ACG, Costa Rica and the world. The collected plant gall morphotypes and their associated organisms from the Guanacaste Conservation Area, were characterized and recorded in a biological database within a specialized herbarium. We named this integrated herbarium a cecidiarium.
When we compare trichome-like structures and lignified external texture in the recorded gall morphotypes between the deciduous and evergreen forest (Bosque Viejo), 55% of the total morphotypes exhibit trichome-like structures and 43.5% has lignified texture in the evergreen forest. The chi-square test did not show significant statistical differences between the two types of forest for these two phenotypic variables (p>0.05, 2 (α =0,.05, df=1) = 3.84).

Discussion
A large number of galling insects have been described in recent years and estimates clearly suggest that the number of species yet to be described could significantly exceed the number of known species (Espírito-Santo & Fernandes 2007). New gall morphotypes have been reported continuously, especially from Brazil, mainly as a consequence of an active community of researchers in that country.
Based on the reviewed literature, it appears that 120 morphotypes described and registered in this study are new records, not only for Costa Rica, but also for the world. Although samples were collected in the rainy season, when plants have a greater amount of green biomass, we estimate that the galls recorded here represent a small portion of the total plant galls present in the Guanacaste Conservation Area. This is based on the higher plant diversity and topographic characteristics of the AGC, as well as the difficulties in searching and obtaining samples from the canopy, particularly in tall trees like those in the Bosque Viejo (old forest). The canopy is the forest stratum with the greatest active growth, and so this is an area where it would be more probable to find galls. Additionally, many galls are hidden inside fruits and other plant organs such as roots (Ley-López et al. 2019), and the high plant density interferes with searching for and finding galls.
Most of the galling species in the Mexican tropical dry forest, for instance, occurred on trees and shrubs, with fewer on herbs and climbing plants (Cuevas-Reyes et al. 2014). For the ACG tropical dry forest, the panorama was similar, based on our results. Moreover, in the tropical dry forest of the Caatinga (Pernambuco, Brazil), most galls were induced on leaves (73.44%), stems (20.31%), and on apical buds (6.25%) (Santos et al. 2011b). These results were similar to data obtained for ACG, where 61% of galls were formed on leaves, followed by stems (17.6%), and on apical buds (6.9%) ( Figure 10D). In contrast to the Caatinga, where the most frequent gall shape was spheroid (32.81%), followed by discoid with 25% (Santos et al. 2011b), for ACG the most frequent gall shape was globular (25.2%), but, the percentage of discoid galls was similar with 18.3% (the second one in frequency). Furthermore, glabrous epidermis shows the highest frequency in both tropical dry forests, with 78% for the Caatinga and 77% for the ACG, surprisingly similar values. Additionally, green was the predominant surface color of galls with 27% for ACG, while it was 73% for the Caatinga (Santos at al. 2011b). Moreover, as in the ACG, for the tropical dry forest of Parque da Sapucaia-Brazil (Costa & Araújo 2019), the plant family with the greatest richness of gall morphotypes was Fabaceae. Similarly, the most affected plant organ was the leaf (82.2%). Otherwise, globular (20.6%) and discoid (13.7%) were the most abundant gall shapes in the aforementioned study, which are comparable with the values for ACG (globular 25.2% and discoid with 18.3%). Glabrous was the predominant external texture with 82.7% and green the most frequent gall color (44.8%) in the tropical dry forest of Parque da Sapucaia, while in the ACG 59% of the galls were glabrous and 27% were green.
The presence of trichomes on gall epidermis ("hairy" or pilose) could be important due to the known properties of these anatomical structures in protecting plants against insect attack (especially the phytophagous insects), temperature regulation, and drought resistance by reducing water loss and moisture retention. Moreover, accumulated evidence suggests that trichomes can absorb UV radiation and reduce the damage by UV-B to photosystem II, preventing stomatal closure (Xiao et al. 2017). Lignins have been associated with plant mechanical support, growth, resistance to insect pests, temperature regulation, and drought tolerance, among other functions (Liu et al.2018). Leaving aside the hypothesis that tissue lignification protects the plant against natural enemies, in galls lignification has also been associated with water conduction (Guedes et al. 2019), as well as protection from UV and oxidative damages generated by excess light exposure and water deficiency (Detoni et al. 2011, Arriola et al. 2018. Our data from ACG shows that only 21% of morphotypes exhibit trichomes on the epidermis while 17.6% of gall morphotypes were lignified, which represents less than expected according to the adaptive advantages attributed to both phenotypic traits. Nevertheless, the lignified epidermis was the most frequent texture in stem galls with 74% ( Figure 10E). Given the function that both anatomical traits could have as a selective response to abiotic conditions, when we compare the deciduous and evergreen forest (e.g. Bosque Viejo), our results contrast with what might be expected, since in the evergreen forest 55% of the total morphotypes showed trichome-like structures and 43.5% of all galls exhibit lignified external texture (p>0.05, 2 (α =0,.05, df=1) = 3.84). Hence, the values for these two phenotypic variables did not show significant statistical differences between the two types of forest. The role of lignified and trichome-covered galls in different biomes of the tropical dry forest are still open questions that should be studied in greater detail.
The low success rate in obtaining adult gall-inducers from a given sample can be explained by the lack of knowledge of insect and gall phenologies. The most practical way of increasing the success rate of rearing adults is probably by collecting the same gall morphotype throughout the year since many galls are seasonal. Furthermore, multiple samples would be needed from each plant species, a task that is frequently not possible because of the time and resources required.
In recent years new gall morphotypes are being reported continuously (Nieves-Aldrey et al. 2008, Goetz et al. 2018, Costa & Araújo 2019. The evidence from these new reports suggests that gall diversity is directly correlated with the richness of plant species in the areas under study (Cuevas-Reyes et al. 2014, Araújo 2017, Coelho et al. 2017. Thus, the previous assumption that gall diversity in tropical regions is less than that in xeric temperate regions is not sustainable considering the biodiversity in tropical ecosystems and the results of the latest studies on the diversity of galls in tropical regions. Cuevas-Reyes et al. (2004) found a significant positive correlation between gallinducing insect species richness and plant species abundance in a Mexican tropical dry forest.
Given the above evidence that future inventories in tropical regions will continue to discover an increasing diversity of gall morphotypes, we proposed organizing the plant galls and associated organisms in a specialized herbarium. This allowed us to have wet and dry collections of plant samples, and preserved specimens of the organisms associated with the galls, in combination with physical records and digital databases providing detailed information about the collected morphotypes, including proper image registration. A detailed repository for the inventory will avoid failures in gall descriptions as well as prevent mixing samples and information from plant galls with other unrelated data. For instance, Ley-López et al. (2019), reported that around 25% of plant specimens previously recorded with galls in the herbarium were ambiguous as to whether the tissue alteration was a gall or not. The specialized herbarium established as a direct consequence of this research represents a unique kind of biological collection different from a traditional plant herbarium, and for this reason the name cecidiarium was suggested as a general concept for summarizing these types of biological records. Considering the diversity of plant galls, this effort represents an important reservoir of germplasm that should be conserved under proper conditions.
Despite the increasing number of new gall morphotypes being reported, some problems still remain beforeachieving a standardized classification. Problems include poor gall morphotype characterization, lack of referenced photos or poor quality images, and the non-existence of a standardized nomenclature. Here, we propose a standardized codification system for plant gall morphotype classification, with the aim of avoiding confusion in the registry of these structures. Gall morphotypes were named using the two first letters from the binomial scientific name of the host plant, followed by the numerical order in which they were discovered (Figures 2-10).
Detailed maps with geographic coordinates, altitudinal distribution, forest cover, and other data are useful tools for understanding the ecological context of gall occurrence. Therefore, georeferenced maps with an appropriate data will allow us to pinpoint and contrast host plant distribution with insect-inducer distribution and habitat ( Figure 11). Figure 11. Spatial distribution of gall morphotypes collected in Área de Conservación Guanacaste (ACG), Guanacaste province, northwest region of Costa Rica. Diamond symbols represent sites were only one gall morphotype was collected. Circle symbols represent cluster sites where two or more gall morphotypes were found; circle diameters vary according to the number of gall morphotypes collected in this area. Larger circles mean that more gall morphotypes were found at a specific area. Cluster size varies between 5-400 linear meters.

Funding
This survey was funded and supported by the Technological Institute of Costa Rica (Instituto Tecnológico de Costa Rica, Vicerrectoría de Investigación y Extensión) by the project number 5402-2160-3101.